In recent years, the number of services and applications using the Web has been higher to increase transmission and reception of Web information. An increase in such Web-related communication traffic makes the network line volume insufficient to slow a service response. Internet communication typically attempts to deliver traffic as equitably as possible irrespective of Web information importance. Both an important traffic and an unimportant traffic are uniformly influenced. Since a response time is secured for a more-important network service, preferential control identifying traffic importance must be done.
There are the following four prior art traffic identification and preferential control methods including the Web.
A first system is that a router identifies the port number of TCP/UDP to deliver it for preferential control. In the system, the router identifies the traffic for preferential control using the port number, the transmitting side IP, and the transmitted side IP included in the header of an IP packet. Typically, a client server type network application uses the specified port number for communication. In the system, the router can identify the type of application, the client, and the server for preferential control.
A second system is that a packet is identified in Precedence field of Type of Service of an IP packet to be preferentially controlled by the router. In the system, the client server type network application adds preferential information to the Type of Service of the packet when a client and a server communicate with each other and the application can specify priority for each data flow.
A third system uses RSVP (Resource Reservation Protocol). In the system, before transmission, a packet needing to secure a communication bandwidth to units configuring a path is transmitted to secure the communication bandwidth of the units on the path for communication. In the system, the application can secure the communication bandwidth for each data flow before transmission so that a response time can be assured for the each data flow.
A fourth system is called packet shaping. The data flow of the client server type network application is identified by the port number, the transmitting side IP and the transmitted side IP on a path between a client and a server. The data flow request interval and network window size are controlled for preferential control for each data flow. In the system, a packet shaper can identify the type of application, the client, and the server for preferential control.
The first and second preferential control systems are described in 2475 of RFC (Request for Comments) in detail. The third system is described in 2205 of the RFC in detail.
The prior art first system identifying a data flow by the port number can identify the data flow only by the type of application.
The Web service typically uses the same port number and cannot identify it for each Web request. Preferential control and bandwidth assurance for each Web cannot be done.
A Web request to a delivery service and a stream service using the Web and a service of a Web business application in which properties such as a traffic volume, a required response time and priority are different are handled to be identical. The traffic of an important Web request is oppressed to slow its response.
The traffic volume of the Web application from the server to the client is larger than that from the client to the server. It is effective to preferentially control the traffic from the server to the client. The Web application uses an unspecified server. It is difficult to realize the preferential control system needing control in the server or on the entire path. For example, in the third system, all units on the path must understand the protocol of the RSVP.
Typically, the communication bandwidth is small on the Internet network where the business network such as LAN accesses the Internet of the provider via a public line. This is a bottleneck in using the Internet. It is effective to perform preferential control of the downward line of the communication path as the bottleneck.
All the prior art first to fourth systems have no idea for preferential control and bandwidth assurance for each Web request. In the provider Internet connection service and the business site connection service, all Web requests of users have the same priority. When a user desires to use a Web service by a quick response and other users fully use the line communication bandwidth, the Web service can be used only by a slow response. A service of the same quality can be only provided at the same charge to a user having a large Web service used amount and a user having a small used amount, which is a kind of inequity.
The present invention has been made to solve the above problems and an object of the present invention is to provide a Web information preferential transfer system in which when the site of a user or a business uses the Internet to receive a Web service, the service can be received with priority according to its quality and charge and the existing Internet network such as a client, a Web server and a path used by the user can be used.